Research And Application Of Benchmarking Evaluation Methods In Aluminum Electrolysis

The current benchmarking management is an important way to help the aluminum electrolysis enterprises gain competitive advantage, which can find out the gap in the enterprise management via the industry benchmark, and establish excellent measures to improve the level of production and operation of enterprises. However, it’s essential to do statistical analysis on aluminum electrolytic industry production parameters by using scientific index synthesis evaluation method, which can help understand enterprise performance objectively.This paper realized the aluminum electrolysis benchmarking management evaluation system with the method of software engineering. The main purpose of this paper is to study the multi index comprehensive evaluation method on the basis of this system. According to the characteristics of key index system of aluminum electrolysis production, which proposed the IMPGA-BP evaluation method. Through the comparison of theory and experiment with other evaluation methods, to design and implement of benchmarking system for aluminum electrolysis. The specific research contents include:First of all, using the idea of software engineering requirements to establish the key indicators of aluminium electrolysis production, such as material consumption index, energy consumption index, yield index and so on; Secondly, according to the characteristics of the key indicators of aluminum electrolysis production, through increasing the population quantity and optimizing the selection operator aimed at the shortcoming of existing genetic algorithms premature convergence, and combine the improved genetic algorithm with BP neural network to build a new evaluation model named IMPGA-BP. Verified by experiment with more accurate evaluation results of key indicators in aluminum electrolysis in the evaluation of the production. Compared IMPGA-BP evaluation method with based on the entropy weight method, distance comprehensive evaluation model, BP neural network evaluation method, based on genetic algorithm and BP neural network in theoretical and experimental contrastion. In the light of the experimental results, analyzing the applicability of each method in the production of aluminum electrolytic key indicators benchmarking evaluation. The results show that the proposed method evaluation results are more accurate.Finally, on the basis of this algorithm and taking the idea of software engineering as a guide. According to the existing software process model summarize the system development process based on the incremental iterative function, and apply them in the aluminum electrolysis benchmarking management system development process. Aluminum electrolysis benchmarking system mainly includes basic information management, management of production targets, benchmarking system for the management, index management, dynamic benchmarking, benchmarking index evaluation.